基于LWOA-LSTM的大容量锂电池SOC估计

doi:10.11930/j.issn.1004-9649.202312106

摘要/Abstract

摘要：

准确预测锂电池荷电状态（SOC）对电池安全运行至关重要，分析在电网不同模式下的SOC更是锂电池全面推广的基础。提出一种基于莱维飞行的鲸鱼优化算法（LWOA）优化长短时记忆神经网络（LSTM），对调频模式下的大容量锂离子电池SOC进行估计。首先，分析LSTM神经网络和LWOA算法，构建LWOA-LSTM模型，进行参数优化；然后，选取调频模式下大容量锂离子电池组实验数据，对数据进行预处理和模型训练；最后，实现调频模式下锂电池的SOC估计。试验结果表明：所构建模型能准确预测锂电池SOC，较WOA-LSTM模型，评估指标RMSE和MAE分别降低了25.55%、28.71%，R²上升了0.76%。

关键词: 荷电状态, 锂电池, 鲸鱼优化算法, 长短时记忆网络, 调频模式

Abstract:

Accurate prediction of the state of charge (SOC) of lithium batteries is crucial for their safe operation, and analyzing the SOC in different power grid modes is the basis for the comprehensive promotion of lithium batteries. This paper proposes a whale optimization algorithm based on Levy flight (LWOA) to optimize long short-term memory neural network (LSTM) for estimating the SOC of large capacity lithium-ion batteries in frequency modulation mode. Firstly, the LSTM neural network and LWOA algorithm are analyzed, and the LWOA-LSTM model is constructed to optimize the parameters. Then, the experimental data of the large capacity lithium-ion battery pack in frequency modulation mode are selected for data preprocessing and model training. Finally, SOC estimation of lithium batteries in frequency modulation mode is achieved. The experimental results show that the constructed model can accurately predict the SOC of lithium batteries. Compared with the WOA-LSTM model, the evaluation indicators RMSE and MAE are reduced by 25.55% and 28.71%, respectively, while R² increases by 0.76%.

Key words: state of charge, lithium batteries, whale optimization algorithm, LSTM, frequency modulation mode

马宏忠, 宣文婧, 朱沐雨, 陈悦林. 基于LWOA-LSTM的大容量锂电池SOC估计[J]. 中国电力, 2024, 57(6): 37-44.

Hongzhong MA, Wenjing XUAN, Muyu ZHU, Yuelin CHEN. SOC Estimation of Large Capacity Lithium Batteries Based on LWOA-LSTM[J]. Electric Power, 2024, 57(6): 37-44.

导出引用管理器 EndNote|Ris|BibTeX

链接本文: https://www.electricpower.com.cn/CN/10.11930/j.issn.1004-9649.202312106

https://www.electricpower.com.cn/CN/Y2024/V57/I6/37

图/表 9

图 1 LSTM单元结构

Fig.1 LSTM unit structure

图 2 基于LWOA-LSTM的SOC预测流程

Fig.2 Flow of SOC prediction based on LWOA-LSTM

图 3 充电曲线提取示意

Fig.3 Extraction of charging curves

图 4 调频工况曲线

Fig.4 Curves of frequency modulation working conditions

表 1 WOA和LWOA对LSTM超参数优化结果

Table 1 WOA and LWOA optimization results for LSTM hyperparameters

网络模型	第1层隐藏层层数	第2层隐藏层层数	学习率	批量大小
WOA-LSTM	94	56	0.0084	57
LWOA-LSTM	50	85	0.0042	28

图 5 3种网络模型的电池SOC预测结果

Fig.5 Prediction results of battery SOC using three network models

图 6 测试集电池SOC预测结果

Fig.6 Battery SOC prediction results of test sets

图 7 2种算法适应度曲线对比

Fig.7 Comparison of fitness curves of two algorithms

表 2 各算法评价指标

Table 2 Evaluation indicators of algorithms

模型	E_RMS	E_MA	E_MAP/%	R²
LSTM	0.0192880	0.0106760	3.5351	0.98412
WOA-LSTM	0.0127524	0.0071967	1.2462	0.98962
LWOA-LSTM	0.0094943	0.0051306	0.3894	0.99714

参考文献 22

1	ZHANG M, YANG D F, DU J X, et al. A review of SOH prediction of Li-ion batteries based on data-driven algorithms[J]. Energies, 2023, 16 (7): 3167. DOI
2	晋殿卫, 顾则宇, 张志宏. 锂电池健康度和剩余寿命预测算法研究[J]. 电力系统保护与控制, 2023, 51 (1): 122- 130.
	JIN Dianwei, GU Zeyu, ZHANG Zhihong. Lithium battery health degree and residual life prediction algorithm[J]. Power System Protection and Control, 2023, 51 (1): 122- 130.
3	李浩, 李浩, 杨扬, 等. 基于改进鲸鱼算法优化GRU的PEMFC老化预测[J/OL]. 中国电机工程学报: 1–14[2024-04-07].https://doi.org/10.13334/j.0258-8013.pcsee.230956.
	LI Hao, LI Hao, YANG Yang, et al. PEMFC aging prediction based on improved whale optimization algorithm optimized GRU[J/OL]. Proceedings of the CSEE: 1–14[2024-04-07].https://doi.org/10.13334/j.0258-8013.pcsee.230956.
4	赵斌, 呼如威, 蒋东方, 等. 高寒高海拔地区微网储能锂电池系统优化设计[J]. 中国电力, 2020, 53 (5): 128- 134.
	ZHAO Bin, HU Ruwei, JIANG Dongfang, et al. Optimized design of lithium battery system for microgrid energy storage in severely cold and high elevation regions[J]. Electric Power, 2020, 53 (5): 128- 134.
5	杨帆. 锂离子电池SOC估计与主动均衡控制策略研究[D]. 杭州: 浙江大学, 2023.
	YANG Fan. Research on SOC estimation and active equalization control strategy of lithium-ion battery[D]. Hangzhou: Zhejiang University, 2023.
6	翟苏巍, 李文云, 周成, 等. 基于改进概率神经网络的储能电池荷电状态估计[J]. 智慧电力, 2024, 52 (2): 94- 100.
	ZHAI Suwei, LI Wenyun, ZHOU Cheng, et al. State-of-charge estimation of energy storage batteries based on modified probabilistic neural networks[J]. Smart Grid, 2024, 52 (2): 94- 100.
7	刘巨, 任羽纶, 易柏年, 等. LSTM-EKF算法实现储能集装箱电芯SOC的优化估计[J]. 电力科学与技术学报, 2024, 39 (2): 198- 206.
	LIU Ju, REN Yulun, YI Bonian, et al. Optimal estimation of cell SOC in energy storage container with LSTM-EKF algorithm[J]. Journal of Electric Power Science and Technology, 2024, 39 (2): 198- 206.
8	廖柯伊. 基于锂离子电池简化电化学模型的自适应死区SOC估计算法[D]. 成都: 西南交通大学, 2021.
	LIAO Keyi. Adaptive dead zone SOC estimation algorithm based on simplified electrochemical model of lithium-ion batteries[D]. Chengdu: Southwest Jiaotong University, 2021.
9	孙国帅, 王靖岳, 武旭东, 等. 基于改进无迹卡尔曼滤波锂电池SOC估计[J]. 电工技术, 2023, (17): 31- 36, 43.
	SUN Guoshuai, WANG Jingyue, WU Xudong, et al. SOC estimation of lithium-ion battery based on improved untraced Kalman filtering[J]. Electric Engineering, 2023, (17): 31- 36, 43.
10	海涛, 范攀龙, 王钧. 基于联合EKF-UKF算法的锂电池SOC预估研究[J]. 电源技术, 2023, 47 (11): 1424- 1428. DOI
	HAI Tao, FAN Panlong, WANG Jun. SOC estimation of lithium battery based on joint EKF-UKF algorithm[J]. Chinese Journal of Power Sources, 2023, 47 (11): 1424- 1428. DOI
11	QAYS M O, BUSWIG Y, HOSSAIN M L, et al. Recent progress and future trends on the state of charge estimation methods to improve battery-storage efficiency: a review[J]. CSEE Journal of Power and Energy Systems, 2022, 8 (1): 105- 114.
12	苏振浩, 李晓杰, 秦晋, 等. 基于BP人工神经网络的动力电池SOC估算方法[J]. 储能科学与技术, 2019, 8 (5): 868- 873.
	SU Zhenhao, LI Xiaojie, QIN Jin, et al. SOC estimation method of power battery based on BP artificial neural network[J]. Energy Storage Science and Technology, 2019, 8 (5): 868- 873.
13	王帅, 马鸿雁, 窦嘉铭, 等. 基于UGOA-BP的锂电池SOC估算[J]. 储能科学与技术, 2022, 11 (1): 258- 264.
	WANG Shuai, MA Hongyan, DOU Jiaming, et al. Estimation of lithium-ion battery state of charge based on UGOA-BP[J]. Energy Storage Science and Technology, 2022, 11 (1): 258- 264.
14	张帅涛, 蒋品群, 宋树祥, 等. 基于注意力机制和CNN-LSTM融合模型的锂电池SOC预测[J/OL]. 电源学报: 1–15[2024-04-07]. http://kns.cnki.net/kcms/detail/12.1420.TM.20220222.1322.005.html.
	ZHANG Shuaitao, JIANG Pinqun, SONG Shuxiang, et al. SOC prediction for lithium battery based on the fusion model of attention mechanism and CNN-LSTM[J/OL]. Journal of Power Supply, 2023, 1–15[2024-04-07]. http://kns.cnki.net/kcms/detail/12.1420.TM.20220222.1322.005.html.
15	陈继斌, 李雯雯, 孙彦玺, 等. 基于卷积-双向长短期记忆网络的电池SOC预测[J]. 电源技术, 2022, 46 (5): 532- 535. DOI
	CHEN Jibin, LI Wenwen, SUN Yanxi, et al. Battery SOC prediction based on convolution bidirectional long term and short term memory network[J]. Chinese Journal of Power Sources, 2022, 46 (5): 532- 535. DOI
16	许德刚, 王再庆, 郭奕欣, 等. 鲸鱼优化算法研究综述[J]. 计算机应用研究, 2023, 40 (2): 328- 336.
	XU Degang, WANG Zaiqing, GUO Yixin, et al. Review of whale optimization algorithm[J]. Application Research of Computers, 2023, 40 (2): 328- 336.
17	陆继翔, 张琪培, 杨志宏, 等. 基于CNN-LSTM混合神经网络模型的短期负荷预测方法[J]. 电力系统自动化, 2019, 43 (8): 131- 137. DOI
	LU Jixiang, ZHANG Qipei, YANG Zhihong, et al. Short-term load forecasting method based on CNN-LSTM hybrid neural network model[J]. Automation of Electric Power Systems, 2019, 43 (8): 131- 137. DOI
18	邹红波, 柴延辉, 杨钦贺, 等. 基于混合ISSA-LSTM的锂离子电池剩余使用寿命预测[J]. 电力系统保护与控制, 2023, 51 (19): 21- 31.
	ZOU Hongbo, CHAI Yanhui, YANG Qinhe, et al. Remaining useful life prediction of lithium-ion batteries based on hybrid ISSA-LSTM[J]. Power System Protection and Control, 2023, 51 (19): 21- 31.
19	REN L, DONG J B, WANG X K, et al. A data-driven auto-CNN-LSTM prediction model for lithium-ion battery remaining useful life[J]. IEEE Transactions on Industrial Informatics, 2021, 17 (5): 3478- 3487. DOI
20	朱元富, 贺文武, 李建兴, 等. 基于Bi-LSTM/Bi-GRU循环神经网络的锂电池SOC估计[J]. 储能科学与技术, 2021, 10 (3): 1163- 1176.
	ZHU Yuanfu, HE Wenwu, LI Jianxing, et al. SOC estimation for Li-ion batteries based on Bi-LSTM and Bi-GRU[J]. Energy Storage Science and Technology, 2021, 10 (3): 1163- 1176.
21	孙玉树, 李宏川, 王波, 等. 电池储能系统SOC神经网络融合估计方法[J]. 湖南大学学报(自然科学版), 2023, 50 (10): 31- 40.
	SUN Yushu, LI Hongchuan, WANG Bo, et al. SOC neural network fusion estimation method for battery energy storage system[J]. Journal of Hunan University (Natural Sciences), 2023, 50 (10): 31- 40.
22	黄晓凡, 李佳瑞, 秦玉文, 等. 基于多尺度CNN和残差LSTM网络的储能电站SOC估计[J/OL]. 电源学报: 1–12[2024-04-07]. http://kns.cnki.net/kcms/detail/12.1420.TM.20231023.1653.014.html.
	HUANG Xiaofan, LI Jiarui, QIN Yuwen, et al. A multi-scale CNN and residual LSTM network method for estimating the state of charge in energy storage system[J]. Journal of Power Supply, 1–12[2024-04-07]. http://kns.cnki.net/kcms/detail/12.1420.TM.20231023.1653.014.html.

[1]	么钟然, 孙丽颖. 考虑线路阻抗的分布式储能SOC均衡控制策略[J]. 中国电力, 2024, 57(9): 238-246.
[2]	邵冲, 胡荣义, 余姣, 王明典. 考虑荷电与储氢状态的风光氢储系统动态控制仿真模型[J]. 中国电力, 2024, 57(7): 109-124.
[3]	朱沐雨, 马宏忠, 郭鹏宇, 宣文婧. 典型调峰/调频工况下储能电池组荷电状态估计[J]. 中国电力, 2024, 57(6): 18-26.
[4]	曹宇, 胡鹏飞, 蔡婉琪, 王曦, 江道灼, 梁一桥. 基于MMC的超级电容与蓄电池混合储能系统及其混合同步控制策略[J]. 中国电力, 2024, 57(6): 78-89.
[5]	徐峰亮, 王克谦, 王文豪, 王鹏, 王文烨, 张帅, 赵凤展. 计及激励型需求响应的低压配电网混合储能优化配置[J]. 中国电力, 2024, 57(6): 90-101.
[6]	薛飞, 李宏强, 李旭涛, 徐恒山. 基于LSTM神经网络的双馈风机控制参数辨识方法[J]. 中国电力, 2023, 56(6): 31-39.
[7]	张宏杰, 陈贵凤, 闫宏伟, 杨晓龙, 侯天仁, 张伟. 基于SMOTE与Bayes优化的LSTM网络变压器故障诊断[J]. 中国电力, 2023, 56(10): 164-170.
[8]	王波, 张占营, 张霄, 席晟哲, 邓锦, 米阳. 直流微电网分布式储能系统精确电流分配策略[J]. 中国电力, 2022, 55(8): 96-103,112.
[9]	陈正雄, 帕孜来·马合木提, 沈玮. 基于改进支持向量回归的IGBT老化预测[J]. 中国电力, 2022, 55(7): 1-10.
[10]	张嘉诚, 夏向阳, 邓子豪, 陈贵全. 储能电站安全参与电网一次调频的优化控制策略[J]. 中国电力, 2022, 55(2): 19-27.
[11]	周鑫, 林镜星, 谢志炜, 张铮, 梁濡铎, 欧祖宏. 基于数据增强及降维方法的配网业扩工程分类模型[J]. 中国电力, 2022, 55(12): 91-97.
[12]	陈崇德, 郭强, 宋子秋, 胡阳. 计及碳收益的风电场混合储能容量优化配置[J]. 中国电力, 2022, 55(12): 22-33.
[13]	彭春华, 刘懿, 孙惠娟. 计及条件风险成本的含EH-CSP电站多源发电系统优化运行[J]. 中国电力, 2022, 55(1): 64-74.
[14]	陈浩, 胡俊杰, 袁海峰, 周华嫣然, 罗魁. 计及配电网拥塞的集群电动汽车参与二次调频方法研究[J]. 中国电力, 2021, 54(12): 162-169.
[15]	丁明, 施建雄, 韩平平, 林子豪, 张宇. 光储系统参与电网调频及调峰的综合控制策略[J]. 中国电力, 2021, 54(1): 116-123,174.